> ## Documentation Index > Fetch the complete documentation index at: https://docs.honeyhive.ai/llms.txt > Use this file to discover all available pages before exploring further. # Retrieve events based on filters ## OpenAPI ````yaml post /events/export openapi: 3.0.3 info: title: HoneyHive API version: 1.0.1 servers: - url: https://api.honeyhive.ai security: - BearerAuth: [] paths: /events/export: post: tags: - Events summary: Retrieve events based on filters operationId: getEvents requestBody: required: true content: application/json: schema: type: object properties: project: type: string description: >- Name of the project associated with the event like `New Project` filters: type: array items: $ref: '#/components/schemas/EventFilter' dateRange: type: object properties: $gte: type: string description: >- ISO String for start of date time filter like `2024-04-01T22:38:19.000Z` $lte: type: string description: >- ISO String for end of date time filter like `2024-04-01T22:38:19.000Z` limit: type: number description: >- Limit number of results to speed up query (default is 1000, max is 7500) page: type: number description: Page number of results (default is 1) required: - project - filters responses: '200': description: Success content: application/json: schema: type: object properties: events: type: array items: $ref: '#/components/schemas/Event' totalEvents: type: number description: Total number of events in the specified filter x-codeSamples: - lang: python label: getEvents source: |- import honeyhive from honeyhive.models import components, operations s = honeyhive.HoneyHive( bearer_auth="", ) res = s.events.get_events(request=operations.GetEventsRequestBody( project='', filters=[ components.EventFilter( field='event_type', value='model', operator=components.Operator.IS, type=components.Type.STRING, ), ], )) if res.object is not None: # handle response pass - lang: typescript label: getEvents source: |- import { HoneyHive } from "honeyhive"; import { Operator, TypeT } from "honeyhive/dist/models/components"; async function run() { const sdk = new HoneyHive({ bearerAuth: "", }); const res = await sdk.events.getEvents({ project: "", filters: [ { field: "event_type", value: "model", operator: Operator.Is, type: TypeT.String, }, ], dateRange: {}, }); if (res.statusCode == 200) { // handle response } } run(); components: schemas: EventFilter: type: object properties: field: type: string description: >- The field name that you are filtering by like `metadata.cost`, `inputs.chat_history.0.content` value: type: string description: The value that you are filtering the field for operator: type: string enum: - is - is not - contains - not contains - greater than description: >- The type of filter you are performing - "is", "is not", "contains", "not contains", "greater than" type: type: string enum: - string - number - boolean - id description: >- The data type you are using - "string", "number", "boolean", "id" (for object ids) example: field: event_type operator: is value: model type: string Event: type: object properties: project_id: type: string description: Name of project associated with the event source: type: string description: Source of the event - production, staging, etc event_name: type: string description: Name of the event event_type: type: string enum: - session - model - tool - chain description: >- Specify whether the event is of "session", "model", "tool" or "chain" type event_id: type: string description: Unique id of the event, if not set, it will be auto-generated session_id: type: string description: >- Unique id of the session associated with the event, if not set, it will be auto-generated parent_id: type: string description: Id of the parent event if nested children_ids: type: array items: type: string description: Id of events that are nested within the event config: type: object additionalProperties: true description: >- Associated configuration JSON for the event - model name, vector index name, etc inputs: type: object additionalProperties: true description: Input JSON given to the event - prompt, chunks, etc properties: chat_history: type: array items: type: object additionalProperties: true description: Messages passed to the model outputs: type: object additionalProperties: true description: Final output JSON of the event error: type: string description: Any error description if event failed start_time: type: number description: UTC timestamp (in milliseconds) for the event start end_time: type: integer description: UTC timestamp (in milliseconds) for the event end duration: type: number description: How long the event took in milliseconds metadata: type: object additionalProperties: true description: Any system or application metadata associated with the event feedback: type: object additionalProperties: true description: Any user feedback provided for the event output metrics: type: object additionalProperties: true description: Any values computed over the output of the event user_properties: type: object additionalProperties: true description: Any user properties associated with the event example: project_id: New Project source: playground session_id: caf77ace-3417-4da4-944d-f4a0688f3c23 event_id: 7f22137a-6911-4ed3-bc36-110f1dde6b66 parent_id: caf77ace-3417-4da4-944d-f4a0688f3c23 event_type: model event_name: Model Completion config: model: gpt-3.5-turbo version: v0.1 - Fork provider: openai hyperparameters: temperature: 0 top_p: 1 max_tokens: 1000 presence_penalty: 0 frequency_penalty: 0 stop: [] 'n': 1 template: - role: system content: |- Answer the user's question only using provided context. Context: {{ context }} - role: user content: '{{question}}' type: chat children_ids: [] inputs: context: Hello world question: What is in the context? chat_history: - role: system content: |- Answer the user's question only using provided context. Context: Hello world - role: user content: What is in the context? outputs: role: assistant content: Hello world error: null start_time: '2024-04-01 22:38:19' end_time: '2024-04-01 22:38:19' duration: 824.8056 metadata: cost: 0.00008 completion_tokens: 23 prompt_tokens: 35 total_tokens: 58 feedback: {} metrics: Answer Faithfulness: 5 Answer Faithfulness_explanation: >- The AI assistant's answer is a concise and accurate description of Ramp's API. It provides a clear explanation of what the API does and how developers can use it to integrate Ramp's financial services into their own applications. The answer is faithful to the provided context. Number of words: 18 user_properties: user: google-oauth2|111840237613341303366 securitySchemes: BearerAuth: type: http scheme: bearer ````